From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754614AbaFYFbz (ORCPT ); Wed, 25 Jun 2014 01:31:55 -0400 Received: from bombadil.infradead.org ([198.137.202.9]:44361 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752370AbaFYFby (ORCPT ); Wed, 25 Jun 2014 01:31:54 -0400 Date: Wed, 25 Jun 2014 07:31:38 +0200 From: Peter Zijlstra To: Rik van Riel Cc: linux-kernel@vger.kernel.org, chegu_vinod@hp.com, mgorman@suse.de, mingo@kernel.org Subject: Re: [PATCH 9/7] sched,numa: remove task_h_load from task_numa_compare Message-ID: <20140625053138.GF6758@twins.programming.kicks-ass.net> References: <1403538095-31256-1-git-send-email-riel@redhat.com> <20140624151454.2a05082e@annuminas.surriel.com> <20140625050735.GB28774@twins.programming.kicks-ass.net> <20140625052144.GH3588@twins.programming.kicks-ass.net> <53AA5D2C.9000401@redhat.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="hI0LKvM4dmXJ5Wy2" Content-Disposition: inline In-Reply-To: <53AA5D2C.9000401@redhat.com> User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --hI0LKvM4dmXJ5Wy2 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Jun 25, 2014 at 01:25:00AM -0400, Rik van Riel wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 >=20 > On 06/25/2014 01:21 AM, Peter Zijlstra wrote: > > On Wed, Jun 25, 2014 at 07:07:35AM +0200, Peter Zijlstra wrote: > >> Shall I merge this into patch 3? > >=20 > > Which gets me the below; which is has a wrong changelog. > >=20 > > task_h_load() already computes the load as seen from the root > > group. effective_load() just does a better (and more expensive) job > > of computing the task movement implications of a move. > >=20 > > So the total effect of this patch shouldn't be very big; regular > > load balancing also only uses task_h_load(), see move_tasks(). > >=20 > > Now, we don't run with preemption disabled, don't run as often, > > etc.., so maybe we can indeed use the more expensive variant just > > fine, but does it really matter? >=20 > In my testing, it appears to make a difference between workloads > converging, and workloads sitting with one last thread stuck on > another node that never gets moved... Fair enough; can you provide a new Changelog that I can paste in? --hI0LKvM4dmXJ5Wy2 Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iQIcBAEBAgAGBQJTql66AAoJEHZH4aRLwOS6XoYP/RqumOoMqFPbPV0ZAdUQFVKe bOw1AbNmmWefr738Ka4wzqmSkRFwM/0Cgif+GJF/x9fcB1Azajyili1XOVu/JokA T+VWfpaR/ZVJWFC3YEvEa3370+j3431V+qROz4tWL7P7/BVDIIfUXpsZx1DXQTag dFsiEJAiMfh8hprKpUWrmr5YitxeL9PpPmR1e7zFgrH5Zat1NigfP9bbs/Zy+uRf DMhKSMGIxIaM1A/HRzFZ/0bkmfZbDK1PODsdyhocsgEG9/yjymAz/iNOaL36DYXB Z9BdZo8tWwn1c3/xXw4npqgagaEmRT9zyo3vtFApsgHTjzOrR8rvfmDTYReevMZY PDnxko0BLWS3olGoUHjO3F2yWDqqOKYlrbjnqHgcq2GEwgfiw7eMki8l70a8OIaQ MYJM6n2KywLsKmFtQssWrJ7q08RjjlLh9/3zVT18X6HoHQVq7rIqVxGSadh9obvZ b7Uv0vB45Cw/YcHYs6ii2pxX8XWhfkxmdp4mCxUtRNtX4g021MIJ9ByncHfE4fXl QptHPi0OxlD8o8ZHPTvfdFQhyeK5uC/Y/zklFnCcr43ekQqTT1ZKiZP7OQDBM597 sqPrNK7c4Zqbi2fF0VBQeUTugakKQDVYiacDfWXAd4MGEsZfmo99UC73q2jipGEm kR0yw59RoO+TkSLqF+2G =m8QX -----END PGP SIGNATURE----- --hI0LKvM4dmXJ5Wy2--